LTAU-FF: Loss Trajectory Analysis for Uncertainty in Atomistic Force Fields
Authors:
- Joshua A. VitaJVJoshua A. Vita
- Amit SamantaASAmit Samanta
- Fei ZhouFZFei Zhou
- Vincenzo LordiVLVincenzo Lordi
Preprint
Abstract
Model ensembles are simple and effective tools for estimating the prediction uncertainty of deep learning atomistic force fields. Despite this, widespread adoption of ensemble-based uncertainty quantification (UQ) techniques is limited by the high computational costs incurred by ensembles during both training and inference. In this work we leverage the cumulative distribution functions (CDFs) of per-sample errors obtained over the course of training to efficiently represent the model ensemble, and couple them with a distance-based similarity search in the model latent space. Using these tools, we develop a simple UQ metric (which we call LTAU) that leverages the strengths of ensemble-based techniques without requiring the evaluation of multiple models during either training or inference. As an initial test, we apply our method towards estimating the epistemic uncertainty in atomistic force fields (LTAU-FF) and demonstrate that it can be easily calibrated to accurately predict test errors on multiple datasets from the literature. We then illustrate the utility of LTAU-FF in two practical applications: 1) tuning the training-validation gap for an example dataset, and 2) predicting errors in relaxation trajectories on the OC20 IS2RS task. Though in this work we focus on the use of LTAU with deep learning atomistic force fields, we emphasize that it can be readily applied to any regression task, or any ensemble-generation technique, to provide a reliable and easy-to-implement UQ metric.Keywords
All Related Versions
This publication has 18 references indexed in Scilit:
- Robust and scalable uncertainty estimation with conformal prediction for machine-learned interatomic potentialsMachine Learning: Science and Technology, 2022
- The long road to calibrated prediction uncertainty in computational chemistryThe Journal of Chemical Physics, 2022
- Machine-learning interatomic potentials for materials scienceActa Materialia, 2021
- Uncertainty quantification in classical molecular dynamicsPhilosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences, 2021
- Neural Network Potential Energy Surfaces for Small Molecules and ReactionsChemical Reviews, 2020
- Machine learning for interatomic potential modelsThe Journal of Chemical Physics, 2020
- Addressing uncertainty in atomistic machine learningPhysical Chemistry Chemical Physics, 2017
- Bayesian uncertainty quantification and propagation in molecular dynamics simulations: A high performance computing frameworkThe Journal of Chemical Physics, 2012
- Gaussian Approximation Potentials: The Accuracy of Quantum Mechanics, without the ElectronsPhysical Review Letters, 2010
- Generalized Neural-Network Representation of High-Dimensional Potential-Energy SurfacesPhysical Review Letters, 2007